Despite advances in identifying defective genes underlying neuropathologies, how these defects underlie symptoms is not known, and this gap is a formidable stumbling block for therapeutics. A case in point is Rett Syndrome (RTT), a severe neurological disease in girls. The disease is due to sporadic mutations in the transcription factor, MeCP2, but why loss of MeCP2 causes neuropathology is enigmatic. Further, RTT holds a unique place in neurological disease because key symptoms are reversible in mice by expressing MeCP2 throughout the brain or just in astrocytes, the prominent glial cell type in brain. The rescue opens the door to therapeutic approaches, but requires a better understanding of what is deficient in RTT and precisely what is rescued upon MeCP2 restoration. Traditional approaches, such as microarray analysis, have focused almost exclusively on individual gene transcript changes, primarily in neurons. This approach has not led to clear answers about the functions of MeCP2 or the cellular basis of the disease, in part due to cellular heterogeneity. It also ignores work indicating a role for astrocytes in contributin to symptoms. In no case is there a molecular benchmark for extent of rescue. Our goal is to attack these issues head on by focusing specifically on rescue of RTT symptoms by astrocytes. Here, we perform a co-expression network analysis, using RNA seq combined with membrane proteomics, on brain and on pure populations of cells sorted from murine brain (aim 1). With an eye towards human-specific therapies, we identify the molecular and cellular consequences of loss and gain of MeCP2 in neural cells from RTT patient IPSCs, and test predictions from these studies in human/mouse xenografts (aim 2). Finally, we test a new hypothesis (aim 3), based on recent preliminary results, that reduced excitatory signaling between astrocytes and neurons may be a functional outcome of the alterations in molecular and membrane properties of these cells (aims 1 and 2).